TABLE 5

Multivariate logistic regression analysis: factors related to 10-year incidence of diabetes

Baseline characteristicsOR (95% CI)P value
Model 1*
 IFG (yes vs. no)5.9 (3.0–11.5)<0.001
 BMI (per 1-SD unit)1.5 (1.1–2.0)0.004
 ln HOMA-IR (per 1-SD unit)1.7 (1.1–2.6)0.009
 ln Sluiter’s Index (per 1-SD unit)0.6 (0.4–0.9)0.007
Model 2
 IFG (yes vs. no)5.5 (2.8–11.0)<0.001
 BMI (per 1-SD unit)1.5 (1.1–2.0)0.010
 ln HOMA-IR (per 1-SD unit)1.6 (1.1–2.5)0.021
 ln Sluiter’s Index (per 1-SD unit)0.7 (0.5–0.9)0.047
Model 3
 IFG (yes vs. no)5.7 (2.8–11.4)<0.001
 BMI (per 1-SD unit)1.5 (1.1–2.1)0.007
 ln HOMA-IR (per 1-SD unit)1.7 (1.1–2.5)0.018
 ln Sluiter’s Index (per 1-SD unit)0.7 (0.5–1.0)0.050
  • *

    * ORs, 95% CIs, and P values were derived from logistic regression analysis, which also included sex and age as covariates and were calculated for a 1-SD unit increase of continuous variables;

  • ORs (95% CI) and P values were derived from logistic regression analysis, which also included sex, age, hypertension, dyslipidemia, and IGT as covariates and were calculated for a 1-SD unit increase of continuous variables;

  • ORs and P values were derived from logistic regression analysis, which also included sex, age, alcohol, smoking, physical activity, family history of diabetes, hypertension, dyslipidemia, hyperuricemia, and IGT as covariates and were calculated for a 1-SD unit increase of continuous variables.